Startups today face many challenges when turning technical ideas into reality:
- Speed: How do you accelerate development cycles?
- Savings: How can you cut costs and increase productivity?
- Scale: How do you take the product from concept to market?
The answer is Model-Based Design—an engineering approach that sets up startups for success by helping them take their products from idea to prototype, and from prototype to production.
“We couldn’t have built the 2008 Tesla® Roadster without MathWorks. It would have taken resources that our new automotive startup company simply did not have. We will continue to rely on MATLAB® and Simulink® to help us make informed design decisions for the next generation of Tesla vehicles.”
Model-Based Design centers on the systematic use of models throughout the development process.
The model serves as:
- A visual representation of the design based on block diagrams and other graphical or textual elements. Models promote understanding of design intent—whether it is data flow or system architecture.
- An executable specification of the design. Models enable simulation of system behavior across multiple domains.
With models, engineers can evaluate design tradeoffs, perform continuous verification and validation, and automatically generate code for hardware implementation.
For startups, the inception of a product starts from an idea, which is then refined and shaped into a design concept defined by a set of requirements—and it rapidly evolves into detailed technical specifications. Building a prototype that can meet these specifications is critical for startups. A prototype demonstrates product value in an early stage, solidifies confidence among internal teams, and helps secure funding from external investors who are looking for early evidence of success. Model-Based Design helps startups quickly go from idea to prototype.
From Idea to Prototype
Quickly Get to Speed
When you start working on an idea, you might be scratching your head over a blank canvas. However, you do not have to start from scratch with Model-Based Design. Simulink and its add-on products provide reference examples and prebuilt blocks to help you get started.
You will find reference examples useful as starting points for your new design. The examples are full-system models built for specific applications, such as insulin pumps, wind farms, package delivery drones, and other applications spanning almost every industry.
As you modify your design to include detailed algorithms and build out full capabilities, you can directly add prebuilt blocks to your design. They are encapsulations of algorithmic modules that have been rigorously tested and fully documented—whether they are signal processing algorithms or control techniques. You can add, combine, or modify these algorithms to suit your design needs.
You can also apply ready-to-use and fully parameterizable blocks for modeling components in the system, such as an electrolyzer powering a green hydrogen production system or a rotor powering a vertical takeoff and landing aircraft.
Voyage Auto, an autonomous driving startup, used reference examples to kick-start its development process.
“We decided to begin with the MATLAB adaptive cruise control (ACC) system example. This example includes a Simulink model that uses MPC to implement an ACC system capable of maintaining a set speed or a set distance from a lead vehicle. Within three days we were running the generated code for the ACC in our vehicle.”
Reduce Development Costs and Time-to-Prototype
You start the design process with many ideas in mind, facing a vast design space and massive uncertainty as you explore possible design options. However, you also often face time, budget, hiring, and other resource constraints as a startup. When you start to formalize and converge on your design choice, it is simply unrealistic to test every option with physical prototypes.
With Model-Based Design, you build and simulate your models as virtual prototypes. You can create massive design studies, evaluate design options, and optimize design performance in a digital design environment—drastically reducing the need to build physical prototypes while mitigating risks of budget overruns.
Ather Energy, an electric scooter startup, used modeling and simulation to accelerate development.
“We had lots of promising ideas, but as a small startup, we did not have the time, money, or people to build prototypes to test each one. With Model-Based Design, we identified and validated the best ideas through simulation, making it possible to deliver a more full-featured scooter in less time. Instead of taking two months to build and test a physical prototype, using our Simscape models, we finished in two weeks.”
Carnegie Clean Energy, a wave energy technology startup, used virtual prototyping and simulation to fix design issues and delivered an operating wave farm.
“As a startup, we cannot afford a development cycle in which we build a system, test, make changes, and retest. It would cost too much and take too long to build even a scale model of our full system. We decided to put the engineering effort into virtual prototyping and simulation so that we get the design right first. Finding and fixing problems via simulation is easier and less expensive than testing on hardware prototypes.”
Focus on Your Design—Not the Code
After you have confirmed a design choice and developed a virtual prototype, how do you implement your design as code running on a physical prototype? You can translate your design to code by manually coding your algorithms, but such an approach involves a lot of steps and can introduce errors and inconsistencies in the process. Changes to the design must be manually implemented in code and it is difficult to establish traceability between your design and your code.
Model-Based Design enables you to automatically generate code from your models. You can go from your design to code running on functional prototypes in days, not months. The generated code is efficient, high-quality, readable, and fully traceable with the design, which means the latest generated code always reflects your most up-to-date design. Code generation is a strong approach for startup software development because it lets your startups focus on the high-level design work.
Ellio, an electric bike startup, sped time-to-prototype by automatically generating control code to target embedded hardware.
“Once we had our controller modeled in Simulink, we created a working prototype in a single day by generating code for the Raspberry Pi® using Simulink Coder™. We estimate that it would have taken at least two weeks to manually code and debug software on the prototype bike, especially since we had no real coding expert on the team.”
Preceyes, a surgical robotics startup, created the world’s first eye surgery robot by implementing its software with automatic code generation.
“With MATLAB and Simulink, I did not have to program a low-level architecture for the controller by myself. As the sole software engineer developing the first release, that was a huge advantage—in fact, I doubt if a single engineer could have done the work otherwise.”
Bigfoot Biomedical, a medical technology startup, developed insulin delivery systems with simulation and automatic code generation.
“Model-Based Design lets us abstract away complexity, so we spend our time on modeling and simulating our systems and algorithms, not constructing and debugging huge programs.”
From Prototype to Production
For startups, developing a functional prototype is instrumental in demonstrating product value to investors, suppliers, and customers. However, for a startup to truly achieve commercial success at scale, it must take the product from a proof-of-concept state (often limited in functionality, quality, and performance) to a production-ready state. Model-Based Design helps startups quickly go from prototype to production.
Model Once, Deploy Everywhere
When you go from prototype to production, there is often a need to change the hardware either to increase performance by leveraging more powerful hardware or to reduce cost in mass production by using more cost-effective and commonly available hardware. The change in hardware requirements is a challenge for startups—integrating software with a different hardware platform not only requires in-house hardware expertise, but also requires changes to the software.
Model-Based Design enables you to decouple your software development from hardware because you can generate portable code from your model to target different hardware, such as C/C++ code for microcontrollers, Verilog/VHDL code for FPGAs/ASICs, structured text for PLCs, or CUDA® code for GPUs. MathWorks partners with major hardware vendors to support hardware integration across these platforms.
Use Simulink to automatically generate portable production code.
The code generation support and hardware integration support make it possible for you to model your design once and deploy it to all supported hardware targets. This means you and your team do not have to become hardware experts and can focus on design work, rather than learning about hardware specifics and recoding your existing algorithms to adapt to a new product.
Stem, an energy storage system startup, used Model-Based Design to decouple control software development from microcontroller hardware.
“Model-Based Design enabled us to develop the controller software before we had hardware. When our first boards came in, all the control algorithms were already in place; five days later we were delivering power with code generated by Embedded Coder.”
Dynisma, a motion simulator startup, scaled the design to target different microcontrollers and hardware systems.
“Despite the fact that we’ve got three different product types out in the field, we’re still able to use the same software, which we deploy to three different controllers and three different systems.”
Minimize Defects and Ensure Quality
A key objective when moving from the prototyping stage to the production stage is reducing defects and ensuring product quality. However, startups often face the risk of identifying errors late in the development process. These errors require significant rework and are time-consuming and costly to fix.
Model-Based Design enables you to continuously verify and validate your design by providing tools for you to perform analyses, checks, and tests in every major phase of your development process—all the way from requirements and early design verification to system integration testing.
With simulation, you front-load your verification efforts by shifting time and resources from physical testing to virtual testing. The “shift left” helps cut testing costs associated with equipment and physical prototypes and can eliminate entire categories of errors before you test the product in real-world conditions. Virtual testing also helps you answer “what if” questions and simulate test scenarios or edge cases that are difficult and sometimes impossible to replicate in a real operating environment.
Supporting the Complete V&V Workflow
| Requirements Traceability | Prevent unintended design behavior |
| Requirements Modeling | Formalize and validate requirements |
| Standards Compliance | Ensure design meets standard guidelines |
| Formal Verification | Prove that the design is robust and meets requirements |
| Component and System Testing | Confirm with simulation-based tests that design meets requirements |
| Back-to-Back Testing | Perform equivalence checking and testing for SIL and PIL |
| Coverage Analysis | Verify that design has been completely tested in MIL, SIL, PIL |
| Automatic Test Generation | Generate tests for coverage analysis, back-to-back testing, etc. |
| Static Code Analysis | Check that code meets standards and free of run-time errors |
| Hardware-in-the-Loop Testing | Test controls by emulating physical systems with real-time target computers |
BPG Motors, an electric motorcycle startup, used simulation-based tests to identify product issues and move the product from prototype to preproduction.
“As the Uno moves from prototype to preproduction, we are expanding our use of Simulink to model and simulate aspects of the Uno that would be too costly, dangerous, or time-consuming to experiment with on the actual hardware.
“Simulink simulations also helped us identify a deficiency with our analog-to-digital converter (ADC). Using some basic ADC blocks in Simulink, we built and simulated a simple model that helped us identify dead spots in our control algorithm that were affecting performance.”
Airnamics, an unmanned aerial system startup, eliminated most software bugs by testing the system virtually before their first flight.
“In flight control system performance, reliability and safety are primary concerns. You cannot take shortcuts, because if you do, you’ll eventually crash.
“Model-Based Design lets us virtually test every part of the system on the ground. Bugs that previously took weeks to identify and repeated flight tests to resolve were fixed in hours via simulation. We found more than 95% of control software bugs before the first flight. The result is a safer, more reliable, and higher quality product.”
Achieve Certification
For startups developing safety-critical applications in industries such as aerospace, automotive, medical devices, and renewable energy, the software in the system not only has to pass rigorous testing, but also must adhere to functional safety standards outlined by international standards organizations or industry working groups. It is a challenge for startups to identify the proper tools to use and the right processes to follow for certification workflows.
Model-Based Design provides tools for you to check whether your model and the code that you generate from it complies with the industry standard.
In addition, IEC Certification Kit provides tool qualification artifacts, certificates, and test suites, and generates traceability matrices. This kit helps you qualify code generation and verification tools, such as Embedded Coder®, HDL Coder™, and the Polyspace® product families, and streamline certification of your embedded systems to ISO® 26262, IEC 61508, EN 50128, ISO 25119, and related standards such as IEC 62304 and EN 50657. Certificates and assessment reports from the certification authority TÜV SÜD are included in the kit for the supported products and standards.
Stem, the energy storage system startup mentioned earlier, also used power systems simulations to pass product tests and achieve IEEE® 1547 certification faster.
“We also achieved certification about 25% faster than usual because of Model-Based Design. PowerStore interacts with the grid, so it must be certified to IEEE 1547 for interconnecting distributed resources with electric power systems, among other standards. We simulated certification tests that our design initially failed, made changes to our controller model, regenerated code, and passed the tests the next day.”
Reuse Design for Next-Gen Products
When you are ready to build upon the initial success of your first product launch, Model-Based Design helps you accelerate development efforts for your next-generation products by enabling the reuse of design models from previous iterations in your new design. You can also easily create and manage design variants when you are scaling the product to reach customers with different needs.
VONSCH, a power electronics equipment company, reused design models to rapidly launch new products with a small engineering team.
“For the FOTO CHARGER product, we reused our MPPT algorithms and numerous Simulink models from FOTO CONTROL. MATLAB and Simulink helped us accelerate research and development by a factor of three while giving us the freedom to switch to another hardware platform if we ever want to.”
Phased Adoption
Even with the potential benefits of using Model-Based Design, startups often consider the risks of adopting a new development process. This is especially true for smaller startups that do not have dedicated staff to pilot a new process and learn new tools.
Successful startups have mitigated this risk by introducing Model-Based Design incrementally. They usually start with a single project, identifying early wins that can be achieved using Model-Based Design versus using the current practice. Successful introduction of Model-Based Design involves taking incremental steps that can help a project along without slowing it down:
- Experiment with a small piece of the project.
- Build on initial modeling success.
- Use models to solve specific design problems.
- Stick with the basics.
- Leverage the experience of MathWorks experts.
To understand the experiences and approaches to adoption for small teams, see the white paper How Engineering Teams Adopt Model-Based Design.
A three-person engineering team at Océ adopted Model-Based Design in one to two weeks with the help of MathWorks Training.
“We had no previous experience with Simulink Coder and Stateflow®. However, within one to two weeks of taking MathWorks Training courses, we could describe very complex scenarios without any difficulty.”
Measure the Return on Investment (ROI)
Adoption of Model-Based Design can lead to significant savings during the system engineering phase, the development phase, and the testing phase. Organizations that adopt Model-Based Design realize savings ranging from 20 to 60%, when compared with traditional methods.
To understand how to quantify expected savings of Model-Based Design over a traditional development approach, see the white paper Measuring the Return on Investment of Model-Based Design.
Vanderhall Motor Works, an electric vehicle startup, adopted Model-Based Design and built an all-electric utility task vehicle (UTV) with a limited crew of engineers in less than a year.
“Normally, it would take an army of coders years to write the software for a vehicle control system. Electric vehicles are in a fast-moving market; if we took a conventional development route, we would still just be dreaming about a product and all of our competitors would beat us to the punch.”
The MathWorks Startup Program
The MathWorks Startup Program offers qualified startups low startup pricing, support from application engineers and technical support, training options in local languages, including 50% off training credits, and co-marketing opportunities to showcase your technology or products. The extensive support and resources from MathWorks are especially helpful for startups that may not have the same level of in-house expertise or resources as larger organizations.
RangeAero, an autonomous cargo helicopter startup, worked with the MathWorks Application Engineering team to adopt Model-Based Design tools and solve complex problems.
“The RangeAero technical team worked together with the applications engineering team at MathWorks to learn to use the tools and apply them to our complex applications. The team from MathWorks was always a call away when we needed help regarding the toolboxes required in setting up workflows.”
Monarch Tractor, an autonomous tractor startup, adopted Model-Based Design and delivered its launch vehicle with the support from the MathWorks Startup Program.
[The] MathWorks Startup Program gave Monarch Tractor a leg up on getting their initial vehicles going, starting to test the architecture with their launch vehicles, and rapidly delivering the first tractors to farmers.”
The MathWorks Accelerator Program
The MathWorks Accelerator Program helps startups in partnering accelerators advance their development. Startups are treated as full commercial customers with technical support and guidance from domain experts while receiving free access to industry-proven software.
Forge, an accelerator in India, partnered with the MathWorks Accelerator Program and enabled its startups to adopt Model-Based Design and technical computing tools in development.
“Leveraging the right technology support is crucial to building the right product and building the product right. At Forge, we enable this capability through partnerships with leading technology players. The partnership with MathWorks has enabled our startups to leverage industry-standard tools such as MATLAB and Simulink.”
Featured in:
Xfinito Biodesigns, a startup company that received incubation support from Dayananda Sagar Entrepreneurship Research & Business Incubation Foundation (DERBI Foundation), used MathWorks support to deliver a novel medical device for treating diabetic neuropathy.
“Because of the technical mentorship program and our experience with MATLAB and Simulink, we could accelerate our intelligent development.”
Whether it is creating renewable energy systems to address the climate challenge; defining the future of mobility on the land, in the air, or sea; or advancing quality of life with new medical devices, startups in many industries have consistently achieved immediate and tangible results with Model-Based Design.
With an incremental approach and support from MathWorks, startups can successfully adopt Model-Based Design and deliver innovations quickly, cost-effectively, and efficiently, scaling from idea to production.
